20 MEMOIRS OP THE NATIONAL ACADEMY OF SCIENCES. 



Coining now to the origin of liimips, fixed or movable, and of spines, the change from herba- 

 ceous to arboreal feeding grounds doubtless affected not oidy the shape of the body, causing it in 

 many cases to be thick and fleshy, but also led to a hypertrophy of the piliferous warts common 

 to all Icpidopterous larva'. The change was ])robably not necessarily due to the stimulus of the 

 visits and attacks of parasitic insects, because the low feeders are, if anything, at the present day 

 at least, more subject to injury from them than arboreal caterpillars. The cau^e was probably 

 more pervasive and a result of a change of the environment, such as is seen in the growth of 

 thorns on desert plants, or the knees of the cypress and other water plants, or the aerial roots of 

 ocbids and other ei)iphytes; and that they may have originated with comparative suddenness 

 seems probable when we bear in mind the aerial roots of corn artificially produced in the lifetime 

 of a single individual; though it should be taken into account that plants are far more ])lastic 

 than animals. 



If the reader will look at the recapitulations 'we Iiave given at the end of the detailed life 

 histories of certain Notodontians, it will be seen that not only are there different adaptive charac- 

 ters in the larval, i)upal, and imaginal stages, but that the larva itself in its different stages is 

 ■wonderfully adapted to different surroundings. 



1. At first some, indeed most, species live socially on the underside of the leaves near where 

 they were born, and thus concealed from observation. Many have glandular hairs, while the 

 tubercles are more or less uniform. 



2. Toward the end of Stage II and in Stage III they feed in exposed situations on the upper 

 side of the leaves, and at the same time ajipears the showy style of ornamentation both as regards 

 colors, hairs, and tubercles, approximating to that of the mature caterpillar, whose life apparently 

 is conditioned by its bright colors and bizarre trappings. 



The smooth-bodied, green larvte of Gluphisia, Nadata, Lophodonta. etc., are the primary forms.' 

 Their shape, coloration, and retired habits ally them biologically to the larva' of the European 

 Fanolis piniperda and other smooth-bodied, green caterpillars with reddish or j-ellowish stripes, 

 ■which feed on trees. These smooth larvre are, however, rare and exceptional, especially in North 

 America. 



But now, owing to a change in the environment, there arose a tendency to the hypertrophy of 

 ;the normal jjiliferous warts, and in the actual life history of the caterpillar the tendency manifests 

 -itself in the third stage of larval life. AVe are inclined to believe (1) that the hypertrophy of certain 

 of the tubercles was eft'ected in a comparatively sudden period in consequence of a comparatively 

 sudden change from herbs to trees, and (2) in response to a sudden exigency; (3) that the spines 

 and stiff, dense spinulated hairs were immediately useful in preventing the attacks of parasitic 

 insects, while (4) the poison glands at the base of the tubercles (in the Attaci, etc.) served to render 

 them distasteful to birds, (5) the bright colors serving as danger signals. 



The Lamarckian factors (1) of change (both direct and indirect) in the milieu, (2) ueed, and 

 ((3) change of habit, and the now generally adopted principle that a change of function induces 

 change in organs* and in some or many cases actually induces the hypertrophy and specialization 

 of what otherwise would be indifferent parts or organs; these factors are all-important in the evo- 

 Jution of the colors, ornaments, and outgrowths from the cuticle of caterpillars.^ 



' I am however inclined, since writing the .ibove, to legard Diitana and Pvgiera as the most primitive forms of 

 Notodontians, the smooth-hodied larv:e of Glupliisia being secondary and adaptive forms. 



■•^ R. Marey : Le transformisme et la physiologie cxpcrimentale, Conrs du College de France. Eevne scieutifique, 

 2" S^rie, iv, 818. (Fnnction makes the organ, especially in the ossoons and mnscnlar systems.) 



See also A. Dohrn: Der t>sprung der Wirbelthiere nnd das Princip dcs Fnuctionswechscls, Leipzig, 1875. 



•' It is possible that the close resemblance of the warts, projecti ons, and spines of certain .nrborcal caterpillars 

 which so closely mimic the spines, leaf scars, and projections of tlie branches or twig'- or plants,. has been bronght 

 about in a way analogous to the production of spots and lines on the body of caterpillars. Darwinians attribute 

 this to the action of "protective mimicry," but this expression rather expresses the result of a series of causes to 

 which wo have endeavored to call attention. The effect of dark and light shades and the light and shade in 

 producing the stripes and bars of caterpillars are comparatively direct and manifest; but how can thorns and other 

 projections on trees and shrubs affect caterpillars directly? Given the origination by hypertrophy of warts and 

 spines, and it is then easy to see that by natural selection caterpillars may have finally become adapted so as to 

 mimic similar vegetable growths. Our object is to endeavor to explain the causes of the primary growth and 

 • development of such projections, i. e., to lay the foundation for the action of natural selection. 



